- Email: [email protected]
Purification and separation of pyrrolizidine alkaloids from Boraginaceae on a polystyrene-divinylbenzene resin
Jowzal of Chromatography, 176 (1979) 274-279 @?_IElsevier Scientific Publishing Company, Amsterdam -
CHROM.
Printed in The Netherlvlds
11,887
Mote
purification
and separaeion
a polystyrene-divinylbenzene HINDRIK
J. HUIZING
of pyrrolizidine
alkaloids
from
Boraginaceae
on
resin
and THEO M. MALINGS
Laboratory for Pharmacognosy (The Netherlands)
and Galenic Pharmacy,
Antonius DeusingIaan 2, 9713 A W Groningen
(Received March 30th. 1979)
Interest in pyrrolizidine alkaloids has increased considerably in recent years owing to their antitumour and hepatotoxic properties I--‘_ Many isolation procedures have been described by a large number of investigators in the screening of various plant materials_ The most commonly used technique is column chromatography with silica oeP9, alumina’O*“, Kieselguhr1z~13 or celIulose1”~15. If the resoistion is 16-18, which has the disadvantage of poor, prep&tive thin-layer chromatography being rather elaborate, or countercurrent distributiorPg cduld be employed. XAD, a macroreticular polystyrene-divinylbenzene copolymer, has shown to be an effective adsorbent of organic molecules from aqueous solutions2G23. No ionexchange or pore exclusion mechanisms are involved’*. Column separations of organic compounds on XAD-2 have been much Iess employed than simple concentration or isoIation procedures”*“. In this paper we describe methods for the purification and separation of monoand diester pyrrolizidine alkaloids from Boraginaceae on XAD-2 columns using gradient elution with acidified methanol-water mixtures. The separation technique was initially developed for an improved isolation of echimidine (I) and symphytine (II) (Fig. I), major alkaloids in Symphytum species, which could not be separated completely by the use of the above column chromatographic methods. The technique proved to be readily adaptable to wider use in the separation of pyrroiizidine alkaloids.
I
Fig. I. Structure of echimidine (I) and symphytine (Ii)_ EXPERIMENTAL
Plant Materials Radix consolidae (Symphyti radix) was supplied by Caesar and Loretz (Ham-
NOTES
275
burg, G.F.R.). Leaves from Cynoglosszrm nerv~sultl Benth. were kindly provided by the Botanical Garden of Groningen (The Netherlands). Isolation of alkaloids Large-scale extraction of alkaloids from Radix consolidae was performed by an ion-exchange method z6. Isolation of alkaloids from leaves of Cynoglossum nerr~osumwas Performed by a Soxhlet methodr6. Serdoxit was used for reduction of alkaloid N-oxides”. As a control in the regeneration procedure on complete removal of dithionite after reduction of Serdoxit, the final aqueous column washings were checked by adding a few drops of the eluate to 1 ml of a 0.01% (w/v) aqueous solution of diazoresorcinol (resazurine; Serva, Heidelberg, G.F.R.). Dithionite in the column effluent caused decolorizatidn or a ’ change from a blue-violet to a red colour, depending on the concentration. Column parameters Three column sizes were used: (A) 15 x 1 cm I.D. with 74-250,um XAD-2 (purchased from Serva as 300IOOO~mbeads, ground and sieved by the authors); (B) 26 x 1.3 cm I.D.; and (C) 61 x 2 cm I.D. with 50-1OO/~m XAD-2 (Serva). Column B was provided with a cooling jacket through which an ethylene glycol-water mixture was passed from a thermostated water-bath. Isocratic separation of alkaloids on column A Before the separation of alkaloids, the column was conditioned by washing with the eluent. Eluents differing in methanol and water contents and adjusted to pH 2.8 and 5.8 were used. Control of eluted fractions Monitoring of the separation of XAD-2 pre-purified echimidine and symphytine using column A (low loading)‘.was achieved by the use of an ISCO UA4 ultraviolet monitor in the transmission mode at a wavelength of 254 nm with a chart speed of
2.5 cm/h. Determination of tailing factors For the determination of the tailing factors of peaks after continuous UV monitoring, the first intercept formed by the perpendicular from the top of a peak towards its base was divided by the second intercept followed by multiplication by io0 (ref. 20). (Tradient separation of alkaloids from XAD-2 columns (B and C) Prior to the elution of alkaloids, the columns were washed with the weakest iuting agent (see below). After draining of approximately 1 ml of methanolic alkaloid .itract into the column, gradient elution was performed. Gradients were established by the addition of a strong eluting mixture conjting of methanol-water-OS &f potassium chloride-hydrochloric acid (pH 2.0) into
NOTES
276
a mixing chamber containing a weak elutin, 0 mixture consisting of methanol-water0.5 M potassium chloride. The latter chamber was connected to the column. The flow of the eluent was adjusted to 0.5 ml/min by a pinch-cock, which was placed after the detector in order to prevent the formation of air bubbles in the detector owing to expansion of compressed air in the eluent. Linear gradient miking chambers Perspex linear gradient mixing chambers were used in order to make it possible to apply pressure on to the top of the solutions used by connection with a cylinder of compressed air. The working pressure did not exceed 0.4 bar. For column C a gradient mixer, consisting of two COMeCted identical glass jars and a peristaltic pump for supplying the gradient to the column, was used. Thin-layer chromatography of alh-aloids isolated from collected fractions To 3-ml fractions eluted from the columns into separate tubes were added 2 ml of saturated sodium carbonate solution (with a few drops of phenolphthalein solution added to the stock solution) and 1 ml of chloroform. After vigorous shaking on a Vortex mixer and after separation of the phases, the upper aqueous phase was aspirated off. If the upper phase was decolorized additional sodium carbonate solution was added. A IO-@ volume of the chloroform extract of twenty fractions was spotted on to a 10 x 20 cm silica gel GFzsl TLC plate (Merck, Darmstadt, G.F.R.), followed by development with chloroform-methanol-25 “/, ammonia solution (85: 14: l)zs using a saturated chamber. Alkaloids were detected with an undiluted modification of Dragendorff’s reagent (Munier) followed by spraying with a 10% (w/v) solution of sodium nitrite 19v30_ Mass spectrometiy A mass spectrum of the isolated alkaloid from Cyrrogiossunr nervosum was recorded on a Finnigan 3300 quadrupole mass spectrometer, equipped with a 6110 data system, using an electron enei_q of 70 eV, an ionizing current of 10O,~cA, and a calibrated temperature of the combined electron-impact-chemicai-ionization source of 210”. RESULTS
AND
DISCUSSION
Neutral pyrrolizidine alkaloids in ammonia solution (pH 9.5) were completely adsorbed to XAD-2 columns and could not be eluted by a basic eluent. Conversion of the alkaloids into the cationic form (NR,H+) by means of an acidic eluent caused rapid and complete removal of alkaloids from the column. The separation power of XAD at isocratic conditions at a pH below 2.5 in aqueous solution appeared to be very poor. At higher pH the separation improved but the retention times increased unacceptably- By adding methanol to the acidic eluent, the retention times became shorter, but tailing of the peaks disturbed the separations. Fig. 2 shows the relationship between tailing and the variation of pH and the methanol content of the eluents. At a higher methanol concentration and at a constant pH the tailing factor decreased. This increased tailing could possibly be induced by
277
NOTES the
stronger
sorption
of the pyrrolizidine
nucleus to the adsorber
because
of the
suppression of the dissociation of the tertiary base. Tailing of peaks could be suppressed by using very low concentrations of alkaloid_ Fig. 3. shows the isocratic separation of XAD-2-pre-purified echimidine and symphytine using column A with an eluent consisting of 50% methanol in water at a PI-I of tailing
3.5 (HCI) at the microgram
level. front
factor
0
TLC
58
28
V-Wlh echim. UV 254
scan
\ 1
1
1
100 WV MeOH in AD %
s
10
fractions
Fig. 2. Influence of methanol concentration and pH (2.8 and 5.8) on peak profile of XAD-2-separated echimidine and symphytine under isocratic conditions. Fig. 3. Isocratic separation of pre-purified echimidine and symphytine. Conditions: column A; eluent, 50% methanol at pH 3.5 (HCI). The trace of the UV adsorbance changes in the eluent has been drawn on the thin-layer chromatogram which shows the alkaloids in the respective column fractions.
In comparison with Kieselgel TLC plates, the elution behaviour of pyrrolizidine alkaloids eluted from XAD columns could be described as being of the “reversedphase” type. The more hydrophilic echimidine was bound less strongly to XAD than was the more hydrophobic alkaloid symphytine, owin, = to the extra hydroxyl group in the former compound. Another means of eliminating tailing in the separation of alkaloids at higher concentrations was to use gradient elution, and a further improvement in separation xas achieved by cooling the column to 1”. This latter inprovement could possibly be explained by the lowering of diffusion leading to an enhancement 3f the sorption of :&aloids. Fig. 4 shows a typical distribution pattern of XAD-2-separated echimidine ‘Fractions 6-14) and symphytine (fractions 14-20) after TLC of collected column ..-actions. In a similar experiment with a different batch of Radix comolidae, echinatine ‘iactions 2-l I), an unknown alkaloid (fractions 12 and 13), echimidine (fractions ‘-25) and symphytine (fractions 27-30) could be isolated. By addition of potassium lloride (0.5 M) to the clncnt with no other changes in the separation conditions the
NOTES
Fig. 4. Separation of 50 mg of a crude Consolida,- G.X?X extract [right) yielding echimidine (e: fractions 6-14) and symphytine (s: fractions 14-20). Conditions: column B; eluent, 50% methanol; gradient descending to pH 2.0 (HCI); total eluent volume, 400 ml.
separation of echimidine and symphytine could be made even more complete, as shown in Fig. 5. Salting out of the alkaloids from the eluent towards the copolymer could be an explanation for this phenomenon. Preparative purification of crude extracts from Cynoglossutn nervomm on column C using a linear gradient consisting of 50% methanol in deionized water with pH decreasing to 2.0 (HCl) with a total gradient vo!ume of 1600 ml at an elution rate
000000000
Fig. 5. Separation as in Fig. 4, but with additional 0.5 M KC1 in the eluent. Upper chromatogram echimidine (fractions 6-14); IoH-erchromatogram. symphytine (fractions 22-32).
NOTES
.279
of 0.4 ml/mm at room temperature gave an almost pure alkaloid in fractions S-13. In its electron-impact mass spectrum a molecular ion was observed at m/e 299, indicating the molecular formula C15HzsN05. The base peak at m/e 138 represented the retronecinejheliotridine-type pyrrolizidine nucleus3r. ACKNOWLEDGEMENTS
The authors thank Dr. A. P. Bruins for performing the mass spectrometry, Mr. J. J. Duits for the illustrations and the workshop of the Department of Pharmacy for the construction of the gradient mixing chambers. REFERENCES
Deinzer, P. Thormon, D. GrilZinand E. Dickinson, Earned_ Mass Speczrom., 5 (1978) 175.C. W. Quak Jr. and H. J. Segall, J. Chromatogr., 150 (1978) 202. M. Kugehnan, Wen-Chih Liu, M. Axelrod, T. J. McBride and K. V. Rao, Lfoydia, 39 (1976) 125. M. M. Ames and G. Powis, J. Chromatogr., 166 (1978) 519.
1 M.
2 3 4 5
T. R. Rajagopalan and V. Batra, Indian1. Cbem., 155 (1977) 494. 6 T. G. Rao and C. K. Atal, Indian J. Pharm., 38 (1976) 156. 7 T. Furuya and K. Araki, Chem. Pharm. Bull.. 16 (1968) 2512. 8 T. Furuya and M. Hikichi, Phytochemiszry. 10 (1971) 2217. 9 A. Ulubelen and S. Doganira, Phytochemistry, 10 (1971) 441. 10 L. H. Zalkow, L. Gelbaum and E. Keinan, Phytochentistry,17 (1978) 172. I1 M. A. Siddiqi, K. A. Suri, 0. P. Suri and C. K. AtaI, Phytuchemisrry, 17 (1978) 2049. 12 C. C. J. Culvenor, Ausr. 1. Chem., 7 (1954) 287. 13 C. C. J. Culvenor, L. J. Drummond and J. R. Price, Aust. J. Chem., 7 (1954) 227. 14 R. B. Bradbury and C. C. J. Culvenor, Aust. J. Cbem., 7 (1954) 378. 15 L. B. Bull, C. C. J. Culvenor and A. T. Dick. in A. Neuberger and E. L. Tatum (Editors), Frontiers of Biology, The Pyrrolizidine Alkaloids, North-Holland, Amsterdam, 1968, p. 27. 16 E. Pedersen, Arch. Pharm. Chem., 3 (1975) 55. 17 A. Ulubelen and F. Ocal, Phytochemistry, 16 (1977) 499. 18 K. A. Suri, R. S. Sawhuey and C. K. Atal, Indian3_ Pharm., 37 (1975) 69. 19 C. C. J. Culvenor, S. R. Johns +d L. W. Smith, Aust. J. Chem., 28 (1975) 2319. 20 M. D. Grieser and 0. J. Pietrryk, Anal. Chem., 45 (1973).1348. 21 A. Tateda and J. S. Fritz, J Chromatogr , 152 (1978) 329.’ 22 V. N. Mallet, G. L. Brun, R. N. MacDonald and K. B&cane, J. Chromafogr., 160 (1978) 81. 23 F. SamueIs Brusse, R. Furst and W. P. van Bexmekom, Pharm. Weekbf., 109 (1974) 921. ’ 24 P. van Rossum and R. G. Webb, J. Chromatugr., 150 (1978) 381. 25 C. H. Chu and D. J. Pietrzyk, Anal. Chenr., 4.6 (1974) 330. 26 J. T. Deagen and M. L. Deinzer, Lloyd& 40 (1977) 395. ! 27 H. J. HuizIng and Th. M. MaIin& J. Cl!romutogr., 173 (1979) 187. 2S R. K. Sharma, G. S. Khajuria and C. K. _itaI, J_ Chromatogr., 19 (1965) 433. 29 E. Stahl, Diinnschicht-Chromatographie,Springer, Berlin, 1967. 30 L. A. Anderson, PIanta Med., 32 (1977) 125. 31 W. Neuner-Jehte, H. Nesvadba and G. Spiteller, Monafsh. Chem., 96 (1265) 321.
CHROM.
Printed in The Netherlvlds
11,887
Mote
purification
and separaeion
a polystyrene-divinylbenzene HINDRIK
J. HUIZING
of pyrrolizidine
alkaloids
from
Boraginaceae
on
resin
and THEO M. MALINGS
Laboratory for Pharmacognosy (The Netherlands)
and Galenic Pharmacy,
Antonius DeusingIaan 2, 9713 A W Groningen
(Received March 30th. 1979)
Interest in pyrrolizidine alkaloids has increased considerably in recent years owing to their antitumour and hepatotoxic properties I--‘_ Many isolation procedures have been described by a large number of investigators in the screening of various plant materials_ The most commonly used technique is column chromatography with silica oeP9, alumina’O*“, Kieselguhr1z~13 or celIulose1”~15. If the resoistion is 16-18, which has the disadvantage of poor, prep&tive thin-layer chromatography being rather elaborate, or countercurrent distributiorPg cduld be employed. XAD, a macroreticular polystyrene-divinylbenzene copolymer, has shown to be an effective adsorbent of organic molecules from aqueous solutions2G23. No ionexchange or pore exclusion mechanisms are involved’*. Column separations of organic compounds on XAD-2 have been much Iess employed than simple concentration or isoIation procedures”*“. In this paper we describe methods for the purification and separation of monoand diester pyrrolizidine alkaloids from Boraginaceae on XAD-2 columns using gradient elution with acidified methanol-water mixtures. The separation technique was initially developed for an improved isolation of echimidine (I) and symphytine (II) (Fig. I), major alkaloids in Symphytum species, which could not be separated completely by the use of the above column chromatographic methods. The technique proved to be readily adaptable to wider use in the separation of pyrroiizidine alkaloids.
I
Fig. I. Structure of echimidine (I) and symphytine (Ii)_ EXPERIMENTAL
Plant Materials Radix consolidae (Symphyti radix) was supplied by Caesar and Loretz (Ham-
NOTES
275
burg, G.F.R.). Leaves from Cynoglosszrm nerv~sultl Benth. were kindly provided by the Botanical Garden of Groningen (The Netherlands). Isolation of alkaloids Large-scale extraction of alkaloids from Radix consolidae was performed by an ion-exchange method z6. Isolation of alkaloids from leaves of Cynoglossum nerr~osumwas Performed by a Soxhlet methodr6. Serdoxit was used for reduction of alkaloid N-oxides”. As a control in the regeneration procedure on complete removal of dithionite after reduction of Serdoxit, the final aqueous column washings were checked by adding a few drops of the eluate to 1 ml of a 0.01% (w/v) aqueous solution of diazoresorcinol (resazurine; Serva, Heidelberg, G.F.R.). Dithionite in the column effluent caused decolorizatidn or a ’ change from a blue-violet to a red colour, depending on the concentration. Column parameters Three column sizes were used: (A) 15 x 1 cm I.D. with 74-250,um XAD-2 (purchased from Serva as 300IOOO~mbeads, ground and sieved by the authors); (B) 26 x 1.3 cm I.D.; and (C) 61 x 2 cm I.D. with 50-1OO/~m XAD-2 (Serva). Column B was provided with a cooling jacket through which an ethylene glycol-water mixture was passed from a thermostated water-bath. Isocratic separation of alkaloids on column A Before the separation of alkaloids, the column was conditioned by washing with the eluent. Eluents differing in methanol and water contents and adjusted to pH 2.8 and 5.8 were used. Control of eluted fractions Monitoring of the separation of XAD-2 pre-purified echimidine and symphytine using column A (low loading)‘.was achieved by the use of an ISCO UA4 ultraviolet monitor in the transmission mode at a wavelength of 254 nm with a chart speed of
2.5 cm/h. Determination of tailing factors For the determination of the tailing factors of peaks after continuous UV monitoring, the first intercept formed by the perpendicular from the top of a peak towards its base was divided by the second intercept followed by multiplication by io0 (ref. 20). (Tradient separation of alkaloids from XAD-2 columns (B and C) Prior to the elution of alkaloids, the columns were washed with the weakest iuting agent (see below). After draining of approximately 1 ml of methanolic alkaloid .itract into the column, gradient elution was performed. Gradients were established by the addition of a strong eluting mixture conjting of methanol-water-OS &f potassium chloride-hydrochloric acid (pH 2.0) into
NOTES
276
a mixing chamber containing a weak elutin, 0 mixture consisting of methanol-water0.5 M potassium chloride. The latter chamber was connected to the column. The flow of the eluent was adjusted to 0.5 ml/min by a pinch-cock, which was placed after the detector in order to prevent the formation of air bubbles in the detector owing to expansion of compressed air in the eluent. Linear gradient miking chambers Perspex linear gradient mixing chambers were used in order to make it possible to apply pressure on to the top of the solutions used by connection with a cylinder of compressed air. The working pressure did not exceed 0.4 bar. For column C a gradient mixer, consisting of two COMeCted identical glass jars and a peristaltic pump for supplying the gradient to the column, was used. Thin-layer chromatography of alh-aloids isolated from collected fractions To 3-ml fractions eluted from the columns into separate tubes were added 2 ml of saturated sodium carbonate solution (with a few drops of phenolphthalein solution added to the stock solution) and 1 ml of chloroform. After vigorous shaking on a Vortex mixer and after separation of the phases, the upper aqueous phase was aspirated off. If the upper phase was decolorized additional sodium carbonate solution was added. A IO-@ volume of the chloroform extract of twenty fractions was spotted on to a 10 x 20 cm silica gel GFzsl TLC plate (Merck, Darmstadt, G.F.R.), followed by development with chloroform-methanol-25 “/, ammonia solution (85: 14: l)zs using a saturated chamber. Alkaloids were detected with an undiluted modification of Dragendorff’s reagent (Munier) followed by spraying with a 10% (w/v) solution of sodium nitrite 19v30_ Mass spectrometiy A mass spectrum of the isolated alkaloid from Cyrrogiossunr nervosum was recorded on a Finnigan 3300 quadrupole mass spectrometer, equipped with a 6110 data system, using an electron enei_q of 70 eV, an ionizing current of 10O,~cA, and a calibrated temperature of the combined electron-impact-chemicai-ionization source of 210”. RESULTS
AND
DISCUSSION
Neutral pyrrolizidine alkaloids in ammonia solution (pH 9.5) were completely adsorbed to XAD-2 columns and could not be eluted by a basic eluent. Conversion of the alkaloids into the cationic form (NR,H+) by means of an acidic eluent caused rapid and complete removal of alkaloids from the column. The separation power of XAD at isocratic conditions at a pH below 2.5 in aqueous solution appeared to be very poor. At higher pH the separation improved but the retention times increased unacceptably- By adding methanol to the acidic eluent, the retention times became shorter, but tailing of the peaks disturbed the separations. Fig. 2 shows the relationship between tailing and the variation of pH and the methanol content of the eluents. At a higher methanol concentration and at a constant pH the tailing factor decreased. This increased tailing could possibly be induced by
277
NOTES the
stronger
sorption
of the pyrrolizidine
nucleus to the adsorber
because
of the
suppression of the dissociation of the tertiary base. Tailing of peaks could be suppressed by using very low concentrations of alkaloid_ Fig. 3. shows the isocratic separation of XAD-2-pre-purified echimidine and symphytine using column A with an eluent consisting of 50% methanol in water at a PI-I of tailing
3.5 (HCI) at the microgram
level. front
factor
0
TLC
58
28
V-Wlh echim. UV 254
scan
\ 1
1
1
100 WV MeOH in AD %
s
10
fractions
Fig. 2. Influence of methanol concentration and pH (2.8 and 5.8) on peak profile of XAD-2-separated echimidine and symphytine under isocratic conditions. Fig. 3. Isocratic separation of pre-purified echimidine and symphytine. Conditions: column A; eluent, 50% methanol at pH 3.5 (HCI). The trace of the UV adsorbance changes in the eluent has been drawn on the thin-layer chromatogram which shows the alkaloids in the respective column fractions.
In comparison with Kieselgel TLC plates, the elution behaviour of pyrrolizidine alkaloids eluted from XAD columns could be described as being of the “reversedphase” type. The more hydrophilic echimidine was bound less strongly to XAD than was the more hydrophobic alkaloid symphytine, owin, = to the extra hydroxyl group in the former compound. Another means of eliminating tailing in the separation of alkaloids at higher concentrations was to use gradient elution, and a further improvement in separation xas achieved by cooling the column to 1”. This latter inprovement could possibly be explained by the lowering of diffusion leading to an enhancement 3f the sorption of :&aloids. Fig. 4 shows a typical distribution pattern of XAD-2-separated echimidine ‘Fractions 6-14) and symphytine (fractions 14-20) after TLC of collected column ..-actions. In a similar experiment with a different batch of Radix comolidae, echinatine ‘iactions 2-l I), an unknown alkaloid (fractions 12 and 13), echimidine (fractions ‘-25) and symphytine (fractions 27-30) could be isolated. By addition of potassium lloride (0.5 M) to the clncnt with no other changes in the separation conditions the
NOTES
Fig. 4. Separation of 50 mg of a crude Consolida,- G.X?X extract [right) yielding echimidine (e: fractions 6-14) and symphytine (s: fractions 14-20). Conditions: column B; eluent, 50% methanol; gradient descending to pH 2.0 (HCI); total eluent volume, 400 ml.
separation of echimidine and symphytine could be made even more complete, as shown in Fig. 5. Salting out of the alkaloids from the eluent towards the copolymer could be an explanation for this phenomenon. Preparative purification of crude extracts from Cynoglossutn nervomm on column C using a linear gradient consisting of 50% methanol in deionized water with pH decreasing to 2.0 (HCl) with a total gradient vo!ume of 1600 ml at an elution rate
000000000
Fig. 5. Separation as in Fig. 4, but with additional 0.5 M KC1 in the eluent. Upper chromatogram echimidine (fractions 6-14); IoH-erchromatogram. symphytine (fractions 22-32).
NOTES
.279
of 0.4 ml/mm at room temperature gave an almost pure alkaloid in fractions S-13. In its electron-impact mass spectrum a molecular ion was observed at m/e 299, indicating the molecular formula C15HzsN05. The base peak at m/e 138 represented the retronecinejheliotridine-type pyrrolizidine nucleus3r. ACKNOWLEDGEMENTS
The authors thank Dr. A. P. Bruins for performing the mass spectrometry, Mr. J. J. Duits for the illustrations and the workshop of the Department of Pharmacy for the construction of the gradient mixing chambers. REFERENCES
Deinzer, P. Thormon, D. GrilZinand E. Dickinson, Earned_ Mass Speczrom., 5 (1978) 175.C. W. Quak Jr. and H. J. Segall, J. Chromatogr., 150 (1978) 202. M. Kugehnan, Wen-Chih Liu, M. Axelrod, T. J. McBride and K. V. Rao, Lfoydia, 39 (1976) 125. M. M. Ames and G. Powis, J. Chromatogr., 166 (1978) 519.
1 M.
2 3 4 5
T. R. Rajagopalan and V. Batra, Indian1. Cbem., 155 (1977) 494. 6 T. G. Rao and C. K. Atal, Indian J. Pharm., 38 (1976) 156. 7 T. Furuya and K. Araki, Chem. Pharm. Bull.. 16 (1968) 2512. 8 T. Furuya and M. Hikichi, Phytochemiszry. 10 (1971) 2217. 9 A. Ulubelen and S. Doganira, Phytochemistry, 10 (1971) 441. 10 L. H. Zalkow, L. Gelbaum and E. Keinan, Phytochentistry,17 (1978) 172. I1 M. A. Siddiqi, K. A. Suri, 0. P. Suri and C. K. AtaI, Phytuchemisrry, 17 (1978) 2049. 12 C. C. J. Culvenor, Ausr. 1. Chem., 7 (1954) 287. 13 C. C. J. Culvenor, L. J. Drummond and J. R. Price, Aust. J. Chem., 7 (1954) 227. 14 R. B. Bradbury and C. C. J. Culvenor, Aust. J. Cbem., 7 (1954) 378. 15 L. B. Bull, C. C. J. Culvenor and A. T. Dick. in A. Neuberger and E. L. Tatum (Editors), Frontiers of Biology, The Pyrrolizidine Alkaloids, North-Holland, Amsterdam, 1968, p. 27. 16 E. Pedersen, Arch. Pharm. Chem., 3 (1975) 55. 17 A. Ulubelen and F. Ocal, Phytochemistry, 16 (1977) 499. 18 K. A. Suri, R. S. Sawhuey and C. K. Atal, Indian3_ Pharm., 37 (1975) 69. 19 C. C. J. Culvenor, S. R. Johns +d L. W. Smith, Aust. J. Chem., 28 (1975) 2319. 20 M. D. Grieser and 0. J. Pietrryk, Anal. Chem., 45 (1973).1348. 21 A. Tateda and J. S. Fritz, J Chromatogr , 152 (1978) 329.’ 22 V. N. Mallet, G. L. Brun, R. N. MacDonald and K. B&cane, J. Chromafogr., 160 (1978) 81. 23 F. SamueIs Brusse, R. Furst and W. P. van Bexmekom, Pharm. Weekbf., 109 (1974) 921. ’ 24 P. van Rossum and R. G. Webb, J. Chromatugr., 150 (1978) 381. 25 C. H. Chu and D. J. Pietrzyk, Anal. Chenr., 4.6 (1974) 330. 26 J. T. Deagen and M. L. Deinzer, Lloyd& 40 (1977) 395. ! 27 H. J. HuizIng and Th. M. MaIin& J. Cl!romutogr., 173 (1979) 187. 2S R. K. Sharma, G. S. Khajuria and C. K. _itaI, J_ Chromatogr., 19 (1965) 433. 29 E. Stahl, Diinnschicht-Chromatographie,Springer, Berlin, 1967. 30 L. A. Anderson, PIanta Med., 32 (1977) 125. 31 W. Neuner-Jehte, H. Nesvadba and G. Spiteller, Monafsh. Chem., 96 (1265) 321.